docs

R/marginal_means.R

@@ -192,7 +192,7 @@ marginal_means <- function(model,
 
     # if type is NULL, we backtransform if relevant
     type_string <- sanitize_type(model = model, type = type, calling_function = "predictions")
-    if (type_string == "linkinv(link)") {
+    if (type_string == "invlink(link)") {
         type_call <- "link"
     } else {
         type_call <- type_string
@@ -387,7 +387,7 @@ marginal_means <- function(model,
     out <- equivalence(out, equivalence = equivalence, df = df, ...)
 
     # after assign draws
-    if (identical(type_string, "linkinv(link)")) {
+    if (identical(type_string, "invlink(link)")) {
         linv <- tryCatch(insight::link_inverse(model), error = function(e) identity)
         out <- backtransform(out, transform = linv)
     }

R/predictions.R

@@ -274,7 +274,7 @@ predictions <- function(model,
 
     # if type is NULL, we backtransform if relevant
     type_string <- sanitize_type(model = model, type = type, calling_function = "predictions")
-    if (type_string == "linkinv(link)") {
+    if (type_string == "invlink(link)") {
         type_call <- "link"
     } else {
         type_call <- type_string
@@ -526,7 +526,7 @@ predictions <- function(model,
     out <- equivalence(out, equivalence = equivalence, df = df, ...)
 
     # after rename to estimate / after assign draws
-    if (identical(type_string, "linkinv(link)")) {
+    if (identical(type_string, "invlink(link)")) {
         linv <- tryCatch(insight::link_inverse(model), error = function(e) identity)
         out <- backtransform(out, transform = linv)
     }

R/print.R

@@ -35,6 +35,7 @@ print.marginaleffects <- function(x,
                                   nrows = getOption("marginaleffects_print_nrows", default = 30),
                                   ncols = getOption("marginaleffects_print_ncols", default = 30),
                                   style = getOption("marginaleffects_print_style", default = "summary"),
+                                  type = getOption("marginaleffects_print_type", default = TRUE),
                                   ...) {
 
 
@@ -243,6 +244,9 @@ print.marginaleffects <- function(x,
     if (ncol(x) <= ncols) {
         cat("Columns:", paste(colnames(x), collapse = ", "), "\n")
     }
+    if (isTRUE(type) && !is.null(attr(x, "type"))) {
+        cat("Type: ", attr(x, "type"), "\n")
+    }
     cat("\n")
 
     print_tail <- attr(x, "print_tail")

R/sanitize_type.R

@@ -11,13 +11,13 @@ sanitize_type <- function(model, type, calling_function = NULL) {
     }
     dict <- type_dictionary
     if (calling_function %in% c("slopes", "comparisons")) {
-        dict <- dict[dict$type != "linkinv(link)", , drop = FALSE]
+        dict <- dict[dict$type != "invlink(link)", , drop = FALSE]
     }
 
-    # fixest: linkinv(link) only supported for glm model
+    # fixest: invlink(link) only supported for glm model
     if (inherits(model, "fixest")) {
         if (!isTRUE(hush(model[["method_type"]]) %in% c("feglm"))) {
-            dict <- dict[dict$type != "linkinv(link)", , drop = FALSE]
+            dict <- dict[dict$type != "invlink(link)", , drop = FALSE]
         }
     }
 

R/type_dictionary.R

@@ -49,7 +49,7 @@ hxlr,scale
 hxlr,density
 ivpml,pr
 ivpml,xb
-fixest,linkinv(link)
+fixest,invlink(link)
 fixest,response
 fixest,link
 hurdle,response
@@ -60,14 +60,14 @@ iv_robust,response
 lm,response
 gam,response
 gam,link
-Gam,linkinv(link)
+Gam,invlink(link)
 Gam,response
 Gam,link
 geeglm,response
 geeglm,link
 glimML,response
 glimML,link
-glm,linkinv(link)
+glm,invlink(link)
 glm,response
 glm,link
 glmerMod,response
@@ -104,7 +104,7 @@ mhurdle,E
 mhurdle,Ep
 mhurdle,p
 mlogit,response
-negbin,linkinv(link)
+negbin,invlink(link)
 negbin,response
 negbin,link
 ols,lp

book/articles/predictions.qmd

@@ -178,7 +178,7 @@ pred <- predictions(modlin)
 mean(pred$estimate)
 ```
 
-Note that in GLM models with a non-linear link function, the default `type` is `linkinv(link)`. This means that predictions are first made on the link scale, averaged, and then back transformed. Thus, the average prediction may not be exactly identical to the average of predictions:
+Note that in GLM models with a non-linear link function, the default `type` is `invlink(link)`. This means that predictions are first made on the link scale, averaged, and then back transformed. Thus, the average prediction may not be exactly identical to the average of predictions:
 
 ```{r}
 mod <- glm(vs ~ hp + am, data = mtcars, family = binomial)
@@ -224,7 +224,7 @@ predictions(
     summarize(AAP = mean(estimate))
 ```
 
-Note that the two results are exactly identical when we specify `type="response"` explicitly. However, they will differ slightly when we leave `type` unspecified, because `marginaleffects` will then automatically make predictions and average on the link scale, before backtransforming (`"linkinv(link)"`):
+Note that the two results are exactly identical when we specify `type="response"` explicitly. However, they will differ slightly when we leave `type` unspecified, because `marginaleffects` will then automatically make predictions and average on the link scale, before backtransforming (`"invlink(link)"`):
 
 ```{r}
 predictions(

man-roxygen/type.R

@@ -2,7 +2,7 @@
 #' 
 #' The `type` argument determines the scale of the predictions used to compute quantities of interest with functions from the `marginaleffects` package. Admissible values for `type` depend on the model object. When users specify an incorrect value for `type`, `marginaleffects` will raise an informative error with a list of valid `type` values for the specific model object. The first entry in the list in that error message is the default type.
 #' 
-#' The `linkinv(link)` is a special type defined by `marginaleffects`. It is available for some (but not all) models and functions. With this link type, we first compute predictions on the link scale, then we use the inverse link function to backtransform the predictions to the response scale. This is useful for models with non-linear link functions as it can ensure that confidence intervals stay within desirable bounds, ex: [0,1] for a logit model. Note that an average of estimates with `type="linkinv(link)"` will not always be equivalent to the average of estimates with `type="response"`.
+#' The `invlink(link)` is a special type defined by `marginaleffects`. It is available for some (but not all) models and functions. With this link type, we first compute predictions on the link scale, then we use the inverse link function to backtransform the predictions to the response scale. This is useful for models with non-linear link functions as it can ensure that confidence intervals stay within desirable bounds, ex: [0,1] for a logit model. Note that an average of estimates with `type="invlink(link)"` will not always be equivalent to the average of estimates with `type="response"`.
 #' 
 #' Some of the most common `type` values are:
 #'